In patients with very-late-stage non-dialysis dependent (NDD) CKD (eGFR <25 ml/min/1.73 m2) the optimal transition of care to renal replacement therapy (RRT, i.e., dialysis or transplantation) is not known. Major uncertainty and significant knowledge gaps have persisted pertaining to differential or individualized transitions of care across different age and demographics and different pre-RRT comorbid conditions and events in several key areas including: (1) the best timing for RRT initiation; (2) the optimal RRT type (dialysis vs. transplant); and in the case of dialysis, the best modality (hemo- vs. peritoneal), format (in-center vs. home), frequency (daily vs. infrequent) and vascular access preparation; (3) the post-RRT impact of pre-RRT comorbid conditions and events including blood pressure and glycemic control, acute kidney injury episodes, and management of CKD specific conditions such as anemia and mineral disorders; and (4) the impact of the above pre-RRT conditions on end-of-life care and decision-making. Given the enormous changes occurring in our health care system and given the high costs of dialysis therapy with persistently poor outcomes, there is an urgent need to answer these important questions related to CKD transitions to RRT. Given the inherent limitations of the USRDS that lacks most core data prior to the RRT transition intercept, we propose a highly innovative linkage approach between the USRDS and two exceptionally rich and large longitudinal databases of very-late-stage NDD-CKD patients, i.e., the national (entire US) Veterans Affairs (VA) database and the regional (Southern California) Kaiser Permanente (KPSC) database, each consisting of millions of people including over 20,000 NDD-CKD patients with eGFR<25 ml/min who transitioned to RRT over the 5-year period 1/1/2008-1/1/2013. These cohorts will also provide annual linkage to projected (2013- 2016) data from over 4,000 incident ESRD patients including several hundred kidney transplant recipients each year for 4 years, hence adding over 16,000 linked patients who will transition to RRT. We hypothesize that (1) a pre-RRT data-driven individualized approach to the transition of care into RRT in very-late-stage CKD is associated with more favorable outcomes including greater survival, fewer hospitalizations and reduced costs, if the decision is based on pre-RRT factors such as clinical and lab variables including the CKD progression rate and comorbid conditions combined with demographics, and (2) that a scoring system derived from these pre-RRT data can determine the timing, preparation and modality of RRT to achieve better outcomes. Upon linking the national VA and regional KPSC data with the USRDS to identify those who have transitioned to dialysis or transplantation, we will examine the predictors of short- (first 6 months) and long-term mortality, hospitalizations and costs by generating pertinent de novo variables including pre-RRT eGFR slope, laboratory data trends and comorbidity indices; and these pre-RRT variables on >36,000 transitioned patients will be linked and reported annually and eventually become available to the USRDS to share with researchers.